This paper is part of a two-part publication that aims to experimentally and numerically evaluate the aerodynamic and mechanical damping of a last stage ST blade at low load operation. A three-stage downscaled steam turbine with a snubbered last stage moving blade LSMB has been tested in the T10MW test facility of Doosan Skoda Power R&D Department in the context of the FLEXTURBINE European project (Flexible Fossil Power Plants for the Future Energy Market through new and advanced Turbine Technologies). Aerodynamic and flutter simulations of different low load conditions have been performed. The acquired data are used to validate the unsteady CFD approach for the prediction of the aerodynamic damping in terms of logarithmic decrement. Numerical results have been achieved through an upgraded version of the URANS CFD solver, selecting appropriate and robust numerical setups for the simulation of very low load conditions, such as increased condenser pressure at the exhaust hood outlet. The numerical methods for blade aerodamping estimation are based on the computation of the unsteady pressure response caused by the row vibration. They are usually classified in time-linearized, harmonic balance and non-linear approaches both in frequency and time domain. The validation of all these methods historically started in the field of aeronautical low-pressure turbines and has been gradually extended to compressor blades and steam turbine rows. For the analysis of a steam turbine last rotor blade operating at strong part load conditions, non-linear methods are recommended as these approaches are able to deal with strong nonlinear phenomena such as shock waves and massive flow separations inside the domain. Experimental data have been used to separate the contributions of mechanical and aerodynamic damping, extrapolating to zero mass flow the total measured damping. Finally, the comparisons between the aerodynamic damping coming from measurements and CFD results have been reported in order to highlight the capability to properly predict the last stage blade flutter stability at low load conditions. Such comparisons confirms the flutter free design of the new snubbered LSMB blade.

The current trend in turbomachinery is pushing forward more and more efficient machines, increasing speeds, reducing components mass and improving their vibrational behaviour. Structural topology optimization is a challenging and promising approach to satisfy all these demands, with a very remarkable economic impact. This approach enables the creation of structures characterized by complex three-dimensional geometries, which are usually difficult or impossible to be produced using traditional manufacturing processes. However, thanks to innovative technologies, as new additive manufacturing techniques, it is now possible to effectively exploit topology optimization to develop innovative components. The aim of this work is to demonstrate the applicability of structural topology optimization techniques in turbomachinery, to improve the dynamic performances and vibrational behaviour of critical components. A 3D mock-up blade geometry based on T106 profile has been designed to reproduce a typical rotor blade in design conditions. The blade has been mounted on a rough disk model, to obtain a rotor blisk in order to ensure a wide design space for the optimization. The optimization has been carried out by applying mean and fluctuating loads coming from a 3D unsteady computation of 1.5stage (stator-rotor-stator) together with the centrifugal stresses. The unsteady loads acting on the rotor skin are due to the wake of the upstream stator and the potential field generated by the downstream stator. A new concept design for the blisk has been developed and the optimized geometry has been compared to the original one to highlight the improvements in terms of mass reduction and improved dynamic behaviour. This paper will confirm the suitability of this approach to turbomachinery components and a prototype of optimized geometry will be ready to be manufactured through innovative additive manufacturing techniques for high resistance alloys.

A validated non-linear uncoupled method for flutter stability analysis was employed to estimate the aerodynamic damping of an HP (High-Pressure) steam turbine blade row. Usually such blade rows are not affected to flutter instability problems, yet an estimation of the aerodynamic damping can be useful for an accurate aeromechanical characterization of these kind of blade rows. The geometry under investigation is a typical steam turbine blade row at design point. Computational aeroelastic analyses are performed on the more relevant modeshape, sampling the nodal diameters, in order to well describe the typical aeroelastic stability curve. The presence of the tip shroud implies a strong mechanical coupling between adjacent blades resulting in complex modeshapes with high frequency, significantly different from those usually analyzed in the flutter analysis. The results in term of aerodynamic damping curves are rather different from the usually sinusoidal shape. This is due to the large variation of the frequency over the analyzed nodal diameters, especially at low nodal diameters range. This results are useful to give a better insight in the aeroelastic response of this type of blades.

This work aims to deepen the understanding of the aerodynamic behavior and the performance of a low pressure steam turbine module. Numerical and experimental results obtained on a three-stage low pressure steam turbine (LPT) module are presented. The selected geometry is representative of the state-of-the-art of low pressure sections for small steam turbines. The test vehicle was designed and operated in different operating conditions with dry and wet steam. Different types of measurements are performed for the global performance estimation of the whole turbine and for the detailed analysis of the flow field. Steady and unsteady CFD analyses have been performed by means of viscous, three-dimensional simulations adopting a real gas, equilibrium steam model. Measured inlet/outlet boundary conditions are used for the computations. The fidelity of the computational setup is proven by comparing computational and experimental results. Main performance curves and span-wise distributions show a good agreement in terms of both shape of curves/distributions and absolute values. Finally, an attempt is done to point out where losses are generated and the physical mechanisms involved are investigated and discussed in details.